Robust Deep Reinforcement Learning for Interference Avoidance in Wideband Spectrum

Aref, Mohamed A.; Jayaweera, Sudharman K.

doi:10.1109/ccaaw.2019.8904887

Cited by 8 publications

(10 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, there are N f c number of fully connected layers for estimating the Q-value associated with each action. Considering that the pooling and fully connected layers only take up 5 − 10% of the computational time [42], their impact on the computational complexity of the CNN can be negligible. Accordingly, the value of F 0 × W o 0 × L o 0 is equal to βl s in the DQLEL framework, where β is the temporal memory depth, and l s represents the length of the state that is equal to (4 + N u ).…”

Section: Computational Complexitymentioning

confidence: 99%

DQLEL: Deep Q-Learning for Energy-Optimized LoS/NLoS UWB Node Selection

Hajiakhondi-Meybodi,

Mohammadi,

Hou

et al. 2021

Preprint

View full text Add to dashboard Cite

Recent advancements in Internet of Things (IoTs) have brought about a surge of interest in indoor positioning for the purpose of providing reliable, accurate, and energy-efficient indoor navigation/localization systems. Ultra Wide Band (UWB) technology has been emerged as a potential candidate to satisfy the aforementioned requirements. Although UWB technology can enhance the accuracy of indoor positioning due to the use of a wide-frequency spectrum, there are key challenges ahead for its efficient implementation. On the one hand, achieving high precision in positioning relies on the identification/mitigation Non Line of Sight (NLoS) links, leading to a significant increase in the complexity of the localization framework. On the other hand, UWB beacons have a limited battery life, which is especially problematic in practical circumstances with certain beacons located in strategic positions. To address these challenges, we introduce an efficient node selection framework to enhance the location accuracy without using complex NLoS mitigation methods, while maintaining a balance between the remaining battery life of UWB beacons. Referred to as the Deep Q-Learning Energy-optimized LoS/NLoS (DQLEL) UWB node selection framework, the mobile user is autonomously trained to determine the optimal pair of UWB beacons to be localized based on the 2-D Time Difference of Arrival (TDoA) framework. The effectiveness of the proposed DQLEL framework is evaluated in terms of the link condition, the deviation of the remaining battery life of UWB beacons, location error, and cumulative rewards. Based on the simulation results, the proposed DQLEL framework significantly outperformed its counterparts across the aforementioned aspects.

show abstract

Section: Computational Complexitymentioning

confidence: 99%

DQLEL: Deep Q-Learning for Energy-Optimized LoS/NLoS UWB Node Selection

Hajiakhondi-Meybodi,

Mohammadi,

Hou

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The conventional RL has certain limitations when dealing with anti‐jamming systems with a large state‐action space. To tackle these challenges, some recent work has proposed using deep reinforcement learning (DRL) [52, 53, 95–100] as shown in Table 4. The DRL is a branch of DL where it uses deep artificial neural networks to enhance the learning operation of the traditional RL.…”

Section: Dl‐based Anti‐jamming Techniquesmentioning

confidence: 99%

“…The sensing matrix along with the chosen communications channel index and an indication whether the communications process was successful or not form the state of the underlying problem. The same problem formulation in [52] is considered in [100], but with new operation parameters and state definitions. These modifications reduce the computational complexity and improve the performance compared to that of [52].…”

Section: Dl‐based Anti‐jamming Techniquesmentioning

confidence: 99%

“…For a better analysis, let us consider the same experiment of Fig. 12 to compare between the performance of the DRL‐based anti‐jamming techniques proposed in [95, 100], namely DQN [95] and DDQN [100], respectively. Fig.…”

Section: Dl‐based Anti‐jamming Techniquesmentioning

confidence: 99%

“…The state of the DQN [95] is a 2D array that maintains a history of actions and their corresponding SINR values. The DDQN [100], on the other hand, utilises spectrum sensing from a 3D array state that contains information about the availability of the channels at different time instants.…”

Section: Dl‐based Anti‐jamming Techniquesmentioning

confidence: 99%

See 2 more Smart Citations